| { | |
| "best_metric": 0.6832635402679443, | |
| "best_model_checkpoint": "hBERTv2_wnli/checkpoint-6", | |
| "epoch": 7.0, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.7351, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5211267605633803, | |
| "eval_loss": 0.7259640693664551, | |
| "eval_runtime": 0.0924, | |
| "eval_samples_per_second": 768.523, | |
| "eval_steps_per_second": 10.824, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7223, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6832635402679443, | |
| "eval_runtime": 0.0917, | |
| "eval_samples_per_second": 774.537, | |
| "eval_steps_per_second": 10.909, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.7189, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4507042253521127, | |
| "eval_loss": 0.711009681224823, | |
| "eval_runtime": 0.0911, | |
| "eval_samples_per_second": 778.974, | |
| "eval_steps_per_second": 10.971, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.633333333333333e-05, | |
| "loss": 0.708, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5352112676056338, | |
| "eval_loss": 0.7058689594268799, | |
| "eval_runtime": 0.0936, | |
| "eval_samples_per_second": 758.69, | |
| "eval_steps_per_second": 10.686, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.566666666666667e-05, | |
| "loss": 0.7032, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5352112676056338, | |
| "eval_loss": 0.6924964189529419, | |
| "eval_runtime": 0.0934, | |
| "eval_samples_per_second": 759.999, | |
| "eval_steps_per_second": 10.704, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.466666666666667e-05, | |
| "loss": 0.6987, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4225352112676056, | |
| "eval_loss": 0.712141215801239, | |
| "eval_runtime": 0.09, | |
| "eval_samples_per_second": 788.982, | |
| "eval_steps_per_second": 11.112, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3666666666666666e-05, | |
| "loss": 0.7109, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5352112676056338, | |
| "eval_loss": 0.692805826663971, | |
| "eval_runtime": 0.093, | |
| "eval_samples_per_second": 763.541, | |
| "eval_steps_per_second": 10.754, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 21, | |
| "total_flos": 553639999438848.0, | |
| "train_loss": 0.7138732274373373, | |
| "train_runtime": 38.2187, | |
| "train_samples_per_second": 830.744, | |
| "train_steps_per_second": 3.925 | |
| } | |
| ], | |
| "max_steps": 150, | |
| "num_train_epochs": 50, | |
| "total_flos": 553639999438848.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |